Accepted Manuscript

Development of low band gap molecular donors with phthalimide terminal groups foruse in solution processed organic solar cells

Ronan San Juan, Abby-Jo Payne, Gregory C. Welch, Ala'a F. Eftaiha

PII: S0143-7208(16)30205-4

DOI: 10.1016/j.dyepig.2016.05.015

Reference: DYPI 5247

To appear in: Dyes and Pigments

Received Date: 23 March 2016

Revised Date: 6 May 2016

Accepted Date: 10 May 2016

Please cite this article as: San Juan R, Payne A-J, Welch GC, Eftaiha AF, Development of low bandgap molecular donors with phthalimide terminal groups for use in solution processed organic solar cells,Dyes and Pigments (2016), doi: 10.1016/j.dyepig.2016.05.015.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Development of low band gap molecular donors with phthalimide terminal groups for use in solution processed organic solar cells

Ronan San Juan,1 Abby-Jo Payne,2 Gregory C. Welch1,2*, Ala’a F. Eftaiha3*

1Department of Chemistry, Dalhousie University, 6274 Coburg Road, P.O. Box 15000, Halifax, Nova Scotia, Canada, B3H 4R2

2Department of Chemistry, University of Calgary, 2500 University Drive, Calgary, Alberta, TN2 1N4 3Department of Chemistry, The Hashemite University, P.O. Box 150459, Zarqa 13115, Jordan

* Corresponding Authors Email: gregory.welch@ucalgary.ca (GCW)

alaa.eftaiha@hu.edu.jo (AFE)

Abstract

The synthesis and characterization of a two novel narrow bandgap ‘donor’ small molecule

semiconductors are reported. The new compounds are based upon the popular DTS(FBT-Th2-

Hexyl)2 donor molecule which has a D2-A-D1-A-D2 architecture with D1 = dithienosilole

(DTS), A = 2-fluorobenzothiazole (FBT), and D2 = hexyl-bi-thiophene (Th2-Hexyl). We have

replaced the D2 hexyl-bi-thiophene unit with electron withdrawing phthalimide units. The new

materials were characterized using a combination of theoretical calculations, UV-visible

spectroscopy, cyclic voltammetry, and thermal analysis. The phthalimide substitution resulted in

an overall stabilization of the highest occupied molecular orbital energy levels, and an increase

in both dipole moment and organic solvent solubility. When paired with PC61BM, organic solar

cells gave surprisingly low power conversion efficiencies. Investigation of the active layer

morphologies revealed large phase segregated domains indicating that phthalimide substitution

renders the donor molecule immiscible with fullerene acceptors.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

Keywords: Organic Solar Cells, Bulk Heterojunction, Solution Processed, Small Molecules,

Electron Donors, Phthalimide

Highlights:

- Synthesis of phthalimide end-capped derivatives of DTS(FBT-Th-Pth-Hexyl)2 - Compounds exhibit higher solubility and deeper HOMO levels - Solar cell devices can achieve higher open circuit voltages - Performance limited by unfavorable morphology with large domain phase separation

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

1. Introduction

Organic photovoltaics (OPV) based on solution-processed small molecules have emerged as

a promising energy conversion technology [1–3]. Significant effort focusing on blends of soluble

small molecular donors with fullerene based acceptors have led to dramatic increase in devices

efficiency over the last several years [4–6]. Innovations in materials design [7–10], active layer

processing [11–14], and device engineering [15–17] have led to power conversion efficiencies

(PCE) reaching beyond 7%. Of particular importance is the design of new donor architectures

that are highly modular and allow for subtle structural modifications to tailor optical,

electrochemical and thermal properties as well as self-assembly tendencies [18]. Fine control of

such properties can lead to significant improvements in both device performance and stability

[14,19]. Some of the most widely studied and best performing small molecule architectures are

comprised of electron-rich donor (D) and electron-poor acceptor (A) organic building blocks.

The research groups of Chen [20] and Bazan [21] have independently developed two related but

different D-A type architectures that have yielded the best performing small molecule based

OPV devices to date.

One of the most successful small molecular donor materials reported so far is one consisting

of a dithienosilole (DTS) core molecule flanked with 2-fluorobenzothiazole (FBT) units and

capped with hexyl-bi-thiophene (Th2-Hexyl) end-groups (DTS(FBT-Th2-Hexyl)2), Scheme 1A)

[7,17,22–25]. This molecule possess a central electron-rich DTS core (D1), electron-poor FBT

acceptor moieties (A), and π-conjugated bithiophene donor units (D2), giving the D2-A-D1-A-

D2 architectural structure. When blended with [6,6]-Phenyl C71 butyric acid methyl ester

(PC71BM), a PCE of 9% has been achieved [15]. It has been reported that structural

modifications to the DTS(FBTTh2)2 molecular has had a tremendous impact on the material

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

properties and device performance. For example, replacing FBT with thiadiazolo pyridine (PT)

moiety increased the material solubility and shifted its optical absorption to lower wavelengths

[26,27]. The pyridine based compound had a poor device performance when poly(3,4-

ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) was used as anode interlayer due

to the susceptibility of the pyridyl N-atom for protonation [28]. Bazan and coworkers [29]

replaced the DTS core with a less electron-rich donor, namely silaindacenodithiophene (SIDT),

which resulted in lowering the highest occupied molecular orbital (HOMO) of the donor

molecule and subsequently increased the open circuit voltage (VOC) of the small donor-fullerene

bulk heterojunction (BHJ) solar cells. Increasing the conjugation length along the molecular

backbone by adding additional (A-D2) units shifted the optical absorption of the compound into

the near-infrared region and increased the material and device thermal stability [30]. Changing

the topology of the small molecule through incorporating additional FBT units diminished the

materials propensity for crystallization, which negatively impacted solar devices leading to poor

device performance [31]. In all cases, modifications were made to the internal building blocks,

while keeping hexyl-bithiophene moiety as end capping unit. We envisioned that replacing the

terminal thiophene moiety with a slightly electron-poor group, viz. phthalimide (Pth), would

increase the electron affinity across the π-conjugated backbone and stabilize the frontier

molecular orbitals, subsequently increasing VOC of the fabricated devices [32–34]. Over the past

few years, small molecules containing phthalimide building blocks have shown potential utility

for organic electronic applications [35,36]. Key advantages include one-step synthesis with a

range of choices for tethering the N-alkyl group and the straightforward coupling using direct

heteroarylation conditions [37,38]. Our recent work has shown that phthalimide-end capped

small molecules exhibited high charge carrier mobility and can tailor optical and physical

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

properties [33,39]. Additionally, bis-imide functional groups are known to direct self-assembly

[40–42].

In this work, we reported on the design and synthesis of a new DTS- FBT based small donor

molecule with phthalimde end-capping unit tethered with octyl (C8H17) and hexyl (C6H13)

groups. The molecular structure of the phthalimide end-capped molecule (DTS(FBT-Th-Pht-R)2)

is depicted in Scheme 1B. Octyl and hexyl side chains were chosen to promote the solubility of

the donor molecules in common solvents used for solution processable devices such as

chlorobenzene and chloroform. The potential utility of DTS(FBT-Th-Pht-R)2 as an electron

donor for solar cell application has been investigated by studying its opto-electronic and thermal

properties. Moreover, its photovoltaic performance in BHJ blends using fullerene as an electron

acceptor was examined and compared with DTS(FBT-Th2-Hexyl)2 -fullerene blends reported in

the literature.

2. Materials and Methods

2.1. Chemicals

Unless otherwise stated, all chemicals were used without further purification. 4-bromophthalic

anhydride and N,N-dimethylacetamide were purchased from TCI Chemicals. n-octylamine, n-

hexylamine, 2-(tributylstannyl)thiophene, anhydrous toluene, pivalic acid and K2CO3 were

purchased from Sigma-Aldrich. Pd(OAc)2 and Pd(PPh3)4 were purchased from Strem Chemicals.

3,3’-di-2-ethylhexylsilylene-2,2’-bithiophene was stannylated with trimethylstannyl chloride in

our labs following standard procedures.

2.2. Methods

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

UV-vis spectroscopy and DSC measurements were carried out as reported previously [43]. The

detailed Materials synthesis and characterization are presented in Supplementary Information

(SI).

2.2.1. Nuclear magnetic resonance (NMR)

NMR spectroscopic data were obtained on a Bruker Avance 300 MHz or Avance 500 MHz

spectrometer at room temperature. Chemical shifts are reported in parts per million (ppm). 19F

NMR spectra were referenced to CCl3F (δ = 0 ppm). 1H NMR spectra were referenced to

residual proton peaks of CDCl3 (δ = 7.27 ppm). 13C NMR spectra were referenced to carbon

peaks of CDCl3 (δ = 77.0 ppm).

2.2.2. Cyclic Voltammetry (CV)

CV measurements were performed using a BASi Cell Stand instrument and BASi Epsilon EC

software. Measurements were performed in a three-electrode, one compartment configuration

equipped with silver/silver chloride electrode, platinum wire, and glassy carbon electrode (3 mm

diameter) as a pseudo reference, counter electrode, and working electrode, respectively. The

glassy carbon electrodes were polished with alumina. The measurements were performed using

0.1 M solution of tetrabutylammoniumhexafluorophosphate (TBAPF6) dissolved in an anhydrous

dichloromethane as a supporting electrolyte. All solutions were purged with nitrogen and then

scanned at varying rates (50-200 mV/s) as-is and at 100 mV/s after the addition of a ferrocene

(Fc) standard. The resulting voltammograms were referenced to the oxidation potential of

Fc/Fc+. The values of the HOMO levels (relative to vacuum) were obtained by comparing the

onset of oxidation to the standard hydrogen electrode (SHE), assuming that the HOMO of

Fc/Fc+ is 4.80 eV below the vacuum level.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

2.2.3. Device Fabrication

Standard organic solar cell devices were made following literature precedence but with slight

modifications.[9] The devices were prepared on cleaned, UV/ozone-treated Corning glass

patterned with ITO, upon which the conductive polymer PEDOT:PSS was spin cast at 5000 rpm

for 60 seconds. The active layers were prepared from solutions of 1 or 2 and PC61BM at a weight

ratio of 6:4 in chlorobenzene at an overall concentration of 35 mg mL-1. Solution preperation

follows that reported by Bazan and coworkers (Adv. Mater, 2012, 24, 3646–3649) and was used

for a direct comparison. The solutions were heated for several hours and residual solids were

quickly filtered prior to casting at 80°C under inert atmosphere (1750 rpm for 60 seconds).

Cathodes were deposited by sequential thermal evaporation of 7.5 nm Ca followed by 100 nm

Al. Device characteristics were measured under illumination by a simulated 100 mW cm-2

AM1.5G light source using a 300 W xenon arc lamp with an AM 1.5 global filter. Solar-

simulator irradiance was calibrated using a standard silicon photovoltaic detector.

2.2.4. Atomic Force Microscopy (AFM)

AFM images were obtained using a Bruker Innova atomic force microscope run in tapping mode

with NCHV-A tips with resonant frequencies ~320 kHz. The AFM images were collected over

20µm×20µm and 5µm×5µm scan areas using a scan rate of 0.75 Hz. A scanning resolution of

256 samples per line. Images were collected using NanoScope Analysis software.

3. Results and discussion

3.1. Computational Analysis

To support our hypothesis that the phthalimide end-capping unit should increase the electron

affinity of the DTS based small molecule, density functional theory (DFT) calculations were

performed to analyze the gas phase properties of the synthesized donor molecules and compared

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

with DTS(FBT-Th2)2. For the sake of simplicity, all alkyl chains were truncated to methyl

groups. Figure 1 shows the DFT results of bent and linear conformations of DTS(FBT-Th-Pht)2

and DTS(FBT-Th2)2, respectively, obtained by rotating the FBT unit by 180º in both directions.

In comparison with the bi-thiophene end capping units, the phthalimide groups were more

twisted with respect to the molecular plane, with a torsion angle of 23º compared to 12º angle for

the former. The calculated HOMOs and lowest unoccupied molecular orbitals (LUMOs) of the

two compounds indicated that the electronic distribution were more concentrated at the core of

the DTS(FBT-Th2)2. DTS(FBT-Th-Pth)2 exhibited deeper HOMO- LUMO levels while

maintaining a comparable bandgap. This highlighted the electron-withdrawing nature of the

phthalimide end-capping units.

The conformational geometry of the bent DTS(FBT-Th2)2 was found to be only 1.13 kJ/mol less

than the relative energy of the linear counterpart. It is important to note that the two geometries

were essentially identical in terms of the calculated dipole moments, electronic and optical

properties (vide infra). As a result, it is likely that the molecule exists as a mixture of the

previous two conformations and might adopt other geometries (herein only two conformations

were explored for simplicity and comparison). When looking at the conformations of DTS(FBT-

Th-Pth)2, the linear structure had a lower relative energy by 0.47 kJ/mol compared to the bent

configuration. Like DTS(FBT-Th2)2, the two conformations had nearly identical electronic

properties but exhibited drastically different dipole moments. The dipole moment of the linear

structure was similar in magnitude to DTS(FBT-Th2)2 conformations (~1-2 Debye), but the bent

geometry had a higher dipole moment (7.41 Debye). Such a large difference in the dipole

moment is a good indication that the phthalimide end-capped donor molecule might interact

differently with solvents (solution) and itself (thin-film) compared to DTS(FBT-Th2)2.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

3.2. Materials Synthesis

Scheme 2 shows the synthetic pathway of (5,5’-bis{4-(5-fluorobenzo[c][1,2,5]thiadiazol-7-

yl)thiophen-2-yl)-2-alkylphthalimide}-3,3’-di-2-ethylhexylsilylene-2,2’-bithiophene)

(DTS(FBT-Th-Pht-R)2 1B, 1: alkyl = C8H17 and 2: alkyl = C6H13). Firstly, the alkylphthalimide

group was prepared by reacting 4-bromophthalic anhydride with either octylamine or

hexylamine in gram-scale quantities (typically ~5 g) of the starting materials. The product 5-

bromo-2-alkylphthalimide (Scheme 2A) was purified by recrystallization in isopropanol with

yields greater than 80%. Then, the alkylated phthalimide was subjected to Stille coupling

reaction with tributyltin thiophene in the presence of a palladium catalyst according to literature

precedence [37] to produce thiophene phthalimide (Scheme 2B). Afterwards, thiophene

phthalimide was coupled to dibromofluorobenzene thiadiazole (FBT) via direct heteroarylation

(DHA) cross-coupling procdure to yield 4-bromo(5-fluorobenzo[c][1,2,5]thiadiazol-7-

yl)thiophen-2-yl)-2-alkylphthalimide (Scheme 2C). This DHA step proved to be a superior

method than Stille coupling as it avoids the installation of a organotin directing group [43].

Subsequent Stille coupling with half equivalent of 7,7́-(4,4-bis(2-ethylhexyl)-4H-silolo(3,2-

b:4,5-b́]dithiophene-2,6-diyl) (DTS) in the presence of Pd(PPh3)4 under microwave conditions

led to the formation of the desired product (Scheme 2D). The final product was purified by flash

column chromatography in dichloromethane with 5% triethylamine to give the small molecule in

suitable yields.

To examine the impact of alkyl-phthalimide end-capping units of the solubility of the DTS-based

compound, 1 and 2 and DTS(FBT-Th2)2 were dissolved in chloroform (CHCl3) at 80 °C, then the

solutions were cold to the room temperature (photographs of the solutions are shown in Figure

S5, SI). All compounds were miscible at 80°C. At room temperature, DTS(FBT-Th2)2 formed

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

aggregates on the side walls of the vial, whereas 1 mostly remained in solution, which highlight

the dramatic difference in solubility between the donor molecules. Absolute solubility was

determined for all three materials with 1 bearing octyl chains having the highest solubility but

both 1 and 2 being significantly more soluble than DTS(FBT-Th2)2 (Table S1).

3.3. Optical Properties

The optical properties of DTS(FBT-Th-Pht-R)2 were probed using UV-visible and

photoluminescence spectroscopy. The absorption profiles of the 1 and 2 derivatives are presented

in Figure 2A. The two compounds exhibit identical absorption characteristics that are similar to

absorption profiles of DTS(FBT-Th2-Hexyl)2 reported in the literature [22]. The solution spectra

displays two dominant bands with absorption maxima at ~395 and ~580 nm. Molar absorptivity

at maximum absorption wavelength for 1 and 2 were determined to be 54000 M-1cm-1 and 59000

M-1cm-1, respectively. These values are comparable to the parent compound DTS(FBT-Th2-

Hexyl)2. Upon transitioning from solution to thin-film, a significant red shift, spectra width

broadening, and the emergence of fine structure in the low energy band is observed. The

absorption onset for both compounds occurs at ~790 nm, corresponding to optical bandgap 1.57

eV. The similarity between the absorption maxima of the phthalimide and bi-thiophene

derivatives indicated that the conjugation length of the two compounds does not change

significantly. Both 1 and 2 exhibit a red-shift of ~30 nm in the maximum absorption when

transitioning from solution to film. This is explained by the tendency of molecules to self-

organize when casted from solution and suggest a co-facial π-π stacking order in the solid state.

The emergence of fine structure can be attributed to a more rigid π-conjugated backbone leading

to better defined optical transitions. When examining the photoluminescence data, a Stoke shift

of 95 nm is observed for both materials indicating a certain degree of confirmation change when

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

transitioning from the ground to excited state. The effects of thermal annealing on the neat films

of compounds 1 and 2 are shown in Figure 2c. The optical density of the lowest energy

absorption peak of 1 increases with increasing temperature, indicating that 1 reached a maximum

π-π packing at 130°C or above. In contrast, the lowest energy absorption peak of 2 increases as

the temperature reaches 100°C, but levels off even when annealed to 130°C, indicating that 2

reached its maximum π-π packing when heated to 100°C. This is attributed to the shorter hexyl

chains that allow for better π-π packing than the longer octyl chains of 1.

3.4. Electrochemical Properties

The electrochemical properties of 1 and 2 were measured using cyclic voltammetry (CV).

Figure 3 shows the cyclic voltammograms of the two compounds in dichloromethane (CH2Cl2)

with scan rate of 200 mV/s using ferrocene as an internal reference. The estimated HOMO and

LUMO frontier molecular orbitals energy levels are summarized in Table 1. The voltammograms

of the octyl and hexyl derivatives implies that the two compounds undergo reversible double

oxidation step with a single reduction step. Compounds 1 and 2 have deep HOMO levels of ~ -

5.1 eV, and relatively low lying LUMO levels of ~ -3.2 eV The HOMO-LUMO offset of the two

compounds is in agreement with the optical band gap calculated from solution absorption onset.

As expected, the length of the tethered alkyl chains on the phthalimide end-group did not alter

the electronic properties of the molecule. In comparison with the energy levels of DTS(FBT-Th2-

Hexyl)2 determined from in-house CV measurements (Figure S9), both 1 and 2 have slightly

deeper HOMO energy levels and slightly shallower LUMO energy levels. The lowering of the

HOMO energy levels for 1 and 2 compared to DTS(FBT-Th2-Hexyl)2 is consistent with the DFT

predictions, although the difference is more drastic in the calculations. The calculations do not

accurately predict the differences in the LUMO energy levels. It is noted that there is large error

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

associated with CV data owing to measurement consistency and interpretation of data (i.e.

determination of redox processes onsets), thus for solar cells, the impact of replacing the hexyl-

bi-thiophene end-group with alkyl-phthalimide moiety on the electronic properties of the small

molecule can be better deduced from the photovoltaic parameters (vide infra).

3.5. Thermal Properties

Differential scanning calorimetry (DSC) was used to examine the thermal transitions of the

alkyl-phthalimide derivatives. Both compounds were subjected to three heating-cooling cycles

between 50 and 300°C. The DSC thermograms of the two compounds are shown in Figure 4. 1

exhibits a melting temperature (Tm) of approximately 249°C and a crystallization temperature

(Tc) of approximately 207°C, while those of 2 are approximately 254°C and 217°C, respectively.

The fact that 2 exhibited higher Tm and Tc than 1 indicated that the shorter hexyl chains allow

more π-π intermolecular interactions than the longer octyl chains. In essence, both materials can

tolerate temperatures exceeding 200°C without irreversible changes up to 300°C. This thermal

stability is consistent with other DTS-based donor molecules [27,44–50] and make the alkyl-

phthalimide derivatives suitable for the fabrication of OPV devices.

3.6. OPV-BHJ Device Data

The photovoltaic performance of 1 and 2 were assessed using phenyl-C61-butyric acid methyl

ester (PC61BM) as acceptor material in a conventional BHJ OPV architecture. Indium tin

oxide/poly(3,4-ethylenedioxythiophene:polystyrene sulfonate) (ITO/PEDOT:PSS) was used as

anode and Ca/Al was used as cathode. The active layer was spun from chlorobenzene solution

with a weight ratio of 6:4 donor/PC61BM. Current–voltage curves of as-casted and post-annealed

devices and the average device parameters are presented in Figure 5 and Table 2, respectively.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

13

For the purpose of comparison, a control device made up of 6:4 DTS(FBT-Th2-Hexyl)2/PC61BM

blend was fabricated. The PCE of the control device was ~5.0% with VOC, short circuit current

(JSC) and fill factor (FF)of ~0.800 V, ~10.00 mA cm-2, and ~0.65, respectively.

The as-deposited devices yielded a PCE of 0.3% and 0.08% for 1/PC61BM and 2/PC61BM

blends, respectively. As shown in Table 2, the post-annealed devices comprised of 1 and 2 with

PC61BM displayed higher PCE in comparison with the as-cast ones. On one side, the JSC and FF

of 1/PC61BM devices were increased with increasing the annealing temperature. The maximum

VOC was attained when the active layer was annealed at 120°C and slightly drops thereafter.

These factors resulted in an overall increase in the PCE of the fabricated devices up to 0.7%

when annealed at 200°C. Further annealing to 240°C shows a decrease in PCE. On the other

side, VOC, JSC and FF of the devices fabricated from 2/PC61BM were increased with thermal

annealing up to 200°C (with the exception of a decrease in JSC at 80°C). Once again, further

annealing up to 240°C showed a decrease in PCE. Since the 1 and 2 have quite similar HOMO

level (vide supra), it is anticipated that the different molecular packing arise from alkyl side

chains [51] and morphological changes introduced by film processing have a substantial impact

on the Voc. Moreover, the photovoltaic data indicated that the Voc of the devices incorporated the

octyl-phthalimide based donor was increased by ~ 8% compared to the control device. This

might be explained by the phthalimide electron-withdrawing nature compared to thiophene end-

caps based donor utilized in the control device as indicated by the DFT calculations (vide supra).

The low PCEs obtained from those devices in comparison to that of DTS(FBT-Th2-

Hexyl)2/PC61BM indicated that the replacement of the thiophene moiety with phthalimide group

had a negative impact on the morphology of the fabricated films, which needs to be addressed.

Although different active layer processing strategies would play a significant role in achieving

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

14

high performance solar cells [52–54], the impact of processing condition is beyond the scope of

this study.

3.7. Atomic Force Microscopy

To gain insight into the low photovoltaic performance, atomic force microscopy (AFM) was

used to map the topography of the 1/PC61BM and 2/PC61BM blends and compared to that of the

DTS(FBT-Th2-Hexyl)2/PC61BM blend. Figure 6 shows the topographic images of the as-cast

donor/fullerene blends. While the film comprised of the bi-thiophene based donor is relatively

smooth, zooming in indicates that the features size are suitable for OPV applications. The films

incorporating the phthalimide-based molecules appear rough with large domains. The domains of

2/PC61BM blend are the largest among the blends so it is not surprising devices based on this

active layer blend have the lowest PCEs. It is anticipated that the non-optimal morphology of

1/PC61BM and 2/PC61BM blends hamper charge dissociation and transport, which is a limiting

factor in achieving high PCE values comparable to DTS(FBT-Th2-Hexyl)2/fullerene based

systems. Clearly in the present the case the incorporation of phthalimide end-capping groups

renders the small molecule donor immiscible with the fullerene acceptor, resulting in large

phase-segregated domains. The exact cause of this is under investigation but it is noted that there

are no reports of OPVs using donors with phthalimide end-capping units, thus future results

investigating these systems should be of high interest. In particular the higher organic solvent

solubility of both 1 and 2 opens the door for a wider range of processing options.

4. Conclusions

We have developed a strategy to modify the electronic properties of one of the most successful

small electron donor molecule utilized in bulk-heterojuction organic photovoltaic devices to date.

By replacing the bi-thiophene endcap with alkyl phthalimide, two new donor molecules were

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

15

disclosed. This is the first report of molecular donors bearing phthalimide endcapping units for

organic photovoltaics. These molecules poses higher dipole moments, deeper highest occupied

molecular orbitals, and increased organic solvent solubility that the bi-thiophene analogues.

Incorporation of these molecules into organic bulk heterojunction solar cell devices produced

mixed results. While increases in open circuit voltages were realized for one derivative, large

phase segregated domains were observed which hindered photovoltaic performance. None-the-

less this work has demonstrated the potential for the phthalimide building block to be used for

the construction of materials relevant to organic solar cells.

- Acknowledgements

GCW acknowledges the Canada Research Chairs Program for salary support. AJP is grateful for

a Nova Scotia graduate scholarship. Next Energy Technologies is acknowledged for supporting

RSJ. We are grateful to Professor Ian Hill (Dalhousie Physics) for use of his solar device testing

equipment and ACEnet for computational resources.

- References

[1] Lin Q, Huang H, Jing Y, Fu H, Chang P, Li D, et al. Flexible photovoltaic technologies. J Mater Chem C 2014;2:1233–47. doi:10.1039/C3TC32197E.

[2] Etxebarria I, Ajuria J, Pacios R. Solution-processable polymeric solar cells: A review on materials, strategies and cell architectures to overcome 10%. Org Electron 2015;19:34–60. doi:10.1016/j.orgel.2015.01.014.

[3] Ragoussi M-E, Torres T. New generation solar cells: concepts, trends and perspectives. Chem Commun 2015;51:3957–72. doi:10.1039/C4CC09888A.

[4] Lin Y, Li Y, Zhan X. Small molecule semiconductors for high-efficiency organic photovoltaics. Chem Soc Rev 2012;41:4245–72. doi:10.1039/C2CS15313K.

[5] Roncali J, Leriche P, Blanchard P. Molecular Materials for Organic Photovoltaics: Small is Beautiful. Adv Mater 2014;26:3821–38. doi:10.1002/adma.201305999.

[6] Huang Y, Kramer EJ, Heeger AJ, Bazan GC. Bulk Heterojunction Solar Cells: Morphology and Performance Relationships. Chem Rev 2014;114:7006–43. doi:10.1021/cr400353v.

[7] Sun Y, Welch GC, Leong WL, Takacs CJ, Bazan GC, Heeger AJ. Solution-processed small-molecule solar cells with 6.7% efficiency. Nat Mater 2012;11:44–8. doi:10.1038/nmat3160.

[8] Zhou J, Zuo Y, Wan X, Long G, Zhang Q, Ni W, et al. Solution-Processed and High-Performance Organic Solar Cells Using Small Molecules with a Benzodithiophene Unit. J Am Chem Soc 2013;135:8484–7. doi:10.1021/ja403318y.

[9] Kan B, Zhang Q, Li M, Wan X, Ni W, Long G, et al. Solution-Processed Organic Solar Cells Based on Dialkylthiol-Substituted Benzodithiophene Unit with Efficiency near 10%. J Am Chem Soc 2014;136:15529–32. doi:10.1021/ja509703k.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

16

[10] Jung JW, Russell TP, Jo WH. A Small Molecule Composed of Dithienopyran and Diketopyrrolopyrrole as Versatile Electron Donor Compatible with Both Fullerene and Nonfullerene Electron Acceptors for High Performance Organic Solar Cells. Chem Mater 2015;27:4865–70. doi:10.1021/acs.chemmater.5b01799.

[11] Sun K, Xiao Z, Hanssen E, Klein MFG, Dam HH, Pfaff M, et al. The role of solvent vapor annealing in highly efficient air-processed small molecule solar cells. J Mater Chem A 2014;2:9048–54. doi:10.1039/C4TA01125B.

[12] Perez LA, Rogers JT, Brady MA, Sun Y, Welch GC, Schmidt K, et al. The Role of Solvent Additive Processing in High Performance Small Molecule Solar Cells. Chem Mater 2014;26:6531–41. doi:10.1021/cm5031987.

[13] Li L, Xiao L, Qin H, Gao K, Peng J, Cao Y, et al. High-Efficiency Small Molecule-Based Bulk-Heterojunction Solar Cells Enhanced by Additive Annealing. ACS Appl Mater Interfaces 2015;7:21495–502. doi:10.1021/acsami.5b06691.

[14] Zhang Q, Kan B, Liu F, Long G, Wan X, Chen X, et al. Small-molecule solar cells with efficiency over 9%. Nat Photon 2015;9:35–41.

[15] Kyaw AKK, Wang DH, Wynands D, Zhang J, Nguyen T-Q, Bazan GC, et al. Improved Light Harvesting and Improved Efficiency by Insertion of an Optical Spacer (ZnO) in Solution-Processed Small-Molecule Solar Cells. Nano Lett 2013;13:3796–801. doi:10.1021/nl401758g.

[16] Topple JM, McAfee SM, Welch GC, Hill IG. Pivotal factors in solution-processed, non-fullerene, all small-molecule organic solar cell device optimization. Org Electron 2015;27:197–201. doi:10.1016/j.orgel.2015.09.020.

[17] Wang DH, Kyaw AKK, Gupta V, Bazan GC, Heeger AJ. Enhanced Efficiency Parameters of Solution-Processable Small-Molecule Solar Cells Depending on ITO Sheet Resistance. Adv Energy Mater 2013;3:1161–5. doi:10.1002/aenm.201300277.

[18] Ni W, Wan X, Li M, Wang Y, Chen Y. A-D-A small molecules for solution-processed organic photovoltaic cells. Chem Commun 2015;51:4936–50. doi:10.1039/C4CC09758K.

[19] Sun K, Xiao Z, Lu S, Zajaczkowski W, Pisula W, Hanssen E, et al. A molecular nematic liquid crystalline material for high-performance organic photovoltaics. Nat Commun 2015;6.

[20] Chen Y, Wan X, Long G. High Performance Photovoltaic Applications Using Solution-Processed Small Molecules. Acc Chem Res 2013;46:2645–55. doi:10.1021/ar400088c.

[21] Coughlin JE, Henson ZB, Welch GC, Bazan GC. Design and Synthesis of Molecular Donors for Solution-Processed High-Efficiency Organic Solar Cells. Acc Chem Res 2014;47:257–70. doi:10.1021/ar400136b.

[22] van der Poll TS, Love JA, Nguyen T-Q, Bazan GC. Non-Basic High-Performance Molecules for Solution-Processed Organic Solar Cells. Adv Mater 2012;24:3646–9. doi:10.1002/adma.201201127.

[23] Sharenko A, Proctor CM, van der Poll TS, Henson ZB, Nguyen T-Q, Bazan GC. A High-Performing Solution-Processed Small Molecule:Perylene Diimide Bulk Heterojunction Solar Cell. Adv Mater 2013;25:4403–6. doi:10.1002/adma.201301167.

[24] Kyaw AKK, Wang DH, Gupta V, Zhang J, Chand S, Bazan GC, et al. Efficient Solution-Processed Small-Molecule Solar Cells with Inverted Structure. Adv Mater 2013;25:2397–402. doi:10.1002/adma.201300295.

[25] Ko Kyaw AK, Gehrig D, Zhang J, Huang Y, Bazan GC, Laquai F, et al. High open-circuit voltage small-molecule p-DTS(FBTTh2)2:ICBA bulk heterojunction solar cells - morphology, excited-state dynamics, and photovoltaic performance. J Mater Chem A 2015;3:1530–9. doi:10.1039/C4TA06256F.

[26] Welch GC, Perez LA, Hoven CV, Zhang Y, Dang X-D, Sharenko A, et al. A modular molecular framework for utility in small-molecule solution-processed organic photovoltaic devices. J Mater Chem 2011;21:12700–9. doi:10.1039/C1JM11963J.

[27] Henson ZB, Welch GC, van der Poll T, Bazan GC. Pyridalthiadiazole-Based Narrow Band Gap Chromophores. J Am Chem Soc 2012;134:3766–79. doi:10.1021/ja209331y.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

17

[28] Ratcliff EL, Bakus II RC, Welch GC, van der Poll TS, Garcia A, Cowan SR, et al. Formation of interfacial traps upon surface protonation in small molecule solution processed bulk heterojunctions probed by photoelectron spectroscopy. J Mater Chem C 2013;1:6223–34. doi:10.1039/C3TC31064G.

[29] Love JA, Nagao I, Huang Y, Kuik M, Gupta V, Takacs CJ, et al. Silaindacenodithiophene-Based Molecular Donor: Morphological Features and Use in the Fabrication of Compositionally Tolerant, High-Efficiency Bulk Heterojunction Solar Cells. J Am Chem Soc 2014;136:3597–606. doi:10.1021/ja412473p.

[30] Liu X, Hsu BBY, Sun Y, Mai C-K, Heeger AJ, Bazan GC. High Thermal Stability Solution-Processable Narrow-Band Gap Molecular Semiconductors. J Am Chem Soc 2014;136:16144–7. doi:10.1021/ja510088x.

[31] Lai LF, Love JA, Sharenko A, Coughlin JE, Gupta V, Tretiak S, et al. Topological Considerations for the Design of Molecular Donors with Multiple Absorbing Units. J Am Chem Soc 2014;136:5591–4. doi:10.1021/ja501711m.

[32] Eftaiha AF, Sun J-P, Hendsbee AD, MacAulay C, Hill IG, Welch GC. High open circuit voltage organic solar cells based upon fullerene free bulk heterojunction active layers. Can J Chem 2014;92:932–9.

[33] Hendsbee AD, McAfee SM, Sun J-P, McCormick TM, Hill IG, Welch GC. Phthalimide-based [small pi]-conjugated small molecules with tailored electronic energy levels for use as acceptors in organic solar cells. J Mater Chem C 2015;3:8904–15. doi:10.1039/C5TC01877C.

[34] Josse P, Dalinot C, Jiang Y, Dabos-Seignon S, Roncali J, Blanchard P, et al. Phthalimide end-capped thienoisoindigo and diketopyrrolopyrrole as non-fullerene molecular acceptors for organic solar cells. J Mater Chem A 2016;4:250–6. doi:10.1039/C5TA09171C.

[35] Bloking JT, Han X, Higgs AT, Kastrop JP, Pandey L, Norton JE, et al. Solution-Processed Organic Solar Cells with Power Conversion Efficiencies of 2.5% using Benzothiadiazole/Imide-Based Acceptors. Chem Mater 2011;23:5484–90. doi:10.1021/cm203111k.

[36] Jinnai S, Ie Y, Karakawa M, Aernouts T, Nakajima Y, Mori S, et al. Electron-Accepting π-Conjugated Systems for Organic Photovoltaics: Influence of Structural Modification on Molecular Orientation at Donor–Acceptor Interfaces. Chem Mater 2016. doi:10.1021/acs.chemmater.5b04551.

[37] Hendsbee AD, Macaulay CM, Welch GC. Synthesis of an H-aggregated thiophene–phthalimide based small molecule via microwave assisted direct arylation coupling reactions. Dyes Pigments 2014;102:204–9. doi:10.1016/j.dyepig.2013.10.046.

[38] Areephong J, Hendsbee AD, Welch GC. Facile synthesis of unsymmetrical and [small pi]-extended furan-diketopyrrolopyrrole derivatives through C-H direct (hetero)arylation using a heterogeneous catalyst system. New J Chem 2015:-. doi:10.1039/C5NJ01150G.

[39] Hendsbee AD, Sun J-P, Rutledge LR, Hill IG, Welch GC. Electron deficient diketopyrrolopyrrole dyes for organic electronics: synthesis by direct arylation, optoelectronic characterization, and charge carrier mobility. J Mater Chem A 2014;2:4198–207. doi:10.1039/C3TA14414C.

[40] Wicklein A, Ghosh S, Sommer M, Würthner F, Thelakkat M. Self-Assembly of Semiconductor Organogelator Nanowires for Photoinduced Charge Separation. ACS Nano 2009;3:1107–14. doi:10.1021/nn9001165.

[41] Melucci M, Durso M, Bettini C, Gazzano M, Maini L, Toffanin S, et al. Structure-property relationships in multifunctional thieno(bis)imide-based semiconductors with different sized and shaped N-alkyl ends. J Mater Chem C 2014;2:3448–56. doi:10.1039/C3TC32538E.

[42] Eftaiha AF, Hendsbee AD, Sun J-P, Hill IG. The influence of molecular geometry on photophysical properties and self-assembly of phthalimide end-capped thiophene-based organic molecules. Mater Lett 2015;157:252–5. doi:10.1016/j.matlet.2015.05.136.

[43] Mercier LG, Leclerc M. Direct (Hetero)Arylation: A New Tool for Polymer Chemists. Acc Chem Res 2013;46:1597–605. doi:10.1021/ar3003305.

[44] Darling SB, You F. The case for organic photovoltaics. RSC Adv 2013;3:17633–48. doi:10.1039/C3RA42989J.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

18

[45] Forrest SR. The path to ubiquitous and low-cost organic electronic appliances on plastic. Nature 2004;428:911–8. doi:10.1038/nature02498.

[46] Kaltenbrunner M, White MS, Głowacki ED, Sekitani T, Someya T, Sariciftci NS, et al. Ultrathin and lightweight organic solar cells with high flexibility. Nat Commun 2012;3:770. doi:10.1038/ncomms1772.

[47] Kippelen B, Brédas J-L. Organic photovoltaics. Energy Environ Sci 2009;2:251–61. doi:10.1039/B812502N.

[48] Li G, Zhu R, Yang Y. Polymer solar cells. Nat Photonics 2012;6:153–61. doi:10.1038/nphoton.2012.11.

[49] Søndergaard R, Hösel M, Angmo D, Larsen-Olsen TT, Krebs FC. Roll-to-roll fabrication of polymer solar cells. Mater Today 2012;15:36–49. doi:10.1016/S1369-7021(12)70019-6.

[50] Su Y-W, Lan S-C, Wei K-H. Organic photovoltaics. Mater Today 2012;15:554–62. doi:10.1016/S1369-7021(13)70013-0.

[51] Lan S-C, Chang C-K, Lu Y-H, Lin S-W, Jen AK-Y, Wei K-H. Side chain structure affects the molecular packing and photovoltaic performance of oligothiophene-based solution-processable small molecules. RSC Adv 2015;5:67718–26. doi:10.1039/C5RA12540E.

[52] Dang MT, Wuest JD. Using volatile additives to alter the morphology and performance of active layers in thin-film molecular photovoltaic devices incorporating bulk heterojunctions. Chem Soc Rev 2013;42:9105–26. doi:10.1039/C3CS35447D.

[53] Grob S, Bartynski AN, Opitz A, Gruber M, Grassl F, Meister E, et al. Solvent vapor annealing on perylene-based organic solar cells. J Mater Chem A 2015;3:15700–9. doi:10.1039/C5TA02806J.

[54] Chen H-C, Su Y-W, Wei K-H. A block copolymer enhances the efficiency of small-molecule bulk-heterojunction photovoltaics. J Mater Chem A 2016;4:2228–35. doi:10.1039/C5TA08932H.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

19

- Schemes

Scheme 1

Scheme 2.

B.

A.

p-DTS(FBT-Th2)2

DTS(FBT-Th-Pht-R)2, R = Octyl (1) or Hexyl (2)

N

O

O

R

S

NS

N

Br

F

Pd(OAc)2, 5 mol %

pivalic acid, 30 mol %

K2CO3, 2 eq.

800C, 20 hrs, DMA

Acetic acid, 4hrs Pd(PPh3)4, cat.

Toluene, 1500C, 20 hrs

1/2 eq.

Pd(PPh3)4, cat.

Microwave

1700C, 30 mins, Toluene

H2N-R

B

> 80%

C

40 %

R1: 2-ethylhexyl

1: R = C8H17

2: R = C6H13

70%

A

> 80%

D

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

20

- Figures

Figure 1.

Figure 2.

-0.1

0.2

0.4

0.6

0.8

1.0

300 400 500 600 700 800 900 1000

Nor

mal

ized

Abs

orba

nce

Wavelength (nm)

1 Sol'n2 Sol'n1 Film2 Film

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

300 400 500 600 700 800 900 1000

Nor

mal

ized

Inte

nsity

Nor

mal

ized

Abs

orba

nce

Wavelength (nm)

UV 1PL 1UV 2PL 2

Stokes Shift = 95 nm

-0.6

-0.2

0.2

0.6

1.0

300 400 500 600 700 800 900 1000

Nor

mal

ized

Abs

orba

nce

Wavelength (nm)

As CastA-100A-130

2

1130°C100°C

A. B. C.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

21

Figure 3.

Figure 4.

Figure 5.

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

-2.5 -1.5 -0.5 0.5 1.5

Nor

mal

ized

Cur

rent

Potential vs Ferrocene (V)

1

2

Reduction

Oxidation

-5

-2.5

0

2.5

5

25 125 225 325

Hea

t Flo

w (

mW

)

Temperature (0C)

1

2

cooling

heating

Tc

Tm

-2

-1

0

1

2

3

4

5

-0.5 0 0.5 1 1.5

J (m

A/c

m2 )

Voltage (V)

as-castA-80A-120A-160A-200

-2

-1

0

1

2

3

4

5

-0.5 0 0.5 1 1.5

J (m

A/c

m2 )

Voltage (V)

as-castA-80A-120A-160A-200

A. B.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

22

Figure 6.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

23

- Tables

Table 1.

Small Molecule HOMO (eV) LUMO (eV) Energy Gap (eV) 1 -5.1 -3.2 1.9 2 -5.1 -3.2 1.9

DTS(FBT-Th2-Hexyl)2* -5.0 -3.3 1.7 *Values obtained using our standard electrochemistry practice and are slightly different from those reported in the literature owing to variations in experimental conditions (see: Adv. Mater., 2012, 24, 3646–3649)

Table 2.

1/PC61BM Devices

Annealing Temperature VOC (V) JSC (mA cm-2) FF PCE (%)

As-cast 0.600 ± 0.01 1.34 ± 0.01 0.37 ± 0.01 0.30 ± 0.01

80 °C 0.831 ± 0.01 1.40 ± 0.01 0.40 ± 0.01 0.46 ± 0.01

120 °C 0.874 ± 0.04 1.35 ± 0.09 0.42 ± 0.05 0.49 ± 0.1

160 °C 0.873 ± 0.01 1.54 ± 0.02 0.43 ± 0.01 0.58 ± 0.02

200 °C 0.867 ± 0.01 1.70 ± 0.07 0.47 ± 0.01 0.70 ± 0.04

2/PC61BM Devices

As-cast 0.285 ± 0.01 0.91 ± 0.02 0.31 ± 0.01 0.08 ± 0.01

80 °C 0.547 ± 0.3 0.83 ± 0.02 0.31 ± 0.04 0.14 ± 0.05

120 °C 0.655 ± 0.3 0.86 ± 0.04 0.30 ± 0.06 0.16 ± 0.04

160 °C 0.670 ± 0.2 0.94 ± 0.06 0.30 ± 0.06 0.18 ± 0.01

200 °C 0.734 ± 0.1 1.12 ± 0.04 0.29 ± 0.06 0.24 ± 0.01

DTS(FBT-Th2-Hexyl)2/PC61BM (Control Device)

In-house 0.800 10.0 0.65 5.2

Literature values[22] 0.809 12.8 0.68 7.0

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTCaptions (Schemes, Figures and Tables)

Scheme 1. Chemical structures of: A. Previously reported p-DTS(FBT-Th2-Hexyl)2. B. DTS(FBT-Th-Pht-R)2. Compounds 1 and 2 bearing octyl and hexyl substituted phthalimide end-capping units, respectively.

Scheme 2. Experimental procedure for the synthesis of DTS(FBT-Th-Pht-R)2. 1H, 13C, 19F

and 2D-COSY NMR of 2C and 2D for octyl and hexyl derivatives are shown in found the Supporting Information (SI, Figure S1-S4).

Figure 1. Results obtained from DFT calculations using B3LYP/6-31G(d,p) level of theory comparing bent (left) and linear (right) geometries of DTS(FBT-Th-Pht)2 and DTS(FBT-Th2)2 (upper and lower rows, respectively) including depictions of the optimized structure with truncated alkyl groups, molecular orbital and electronic energy levels descriptions.

Figure 2. A. Normalized UV-vis absorbance spectra of the octyl (1) and hexyl (2) derivatives in CHCl3 solution and thin films spun from CHCl3 solution. B. Excitation and emission spectra of 1 and 2 in CHCl3 solution. C. Absorption profiles of the as-cast and thermally annealed neat films

Figure 3. Cyclic voltammogram of octyl (1) and hexyl (2) derivatives obtained in CH2Cl2 solution under a N2 atmosphere using a sweep rate of 200 mV/s. E(HOMO) = - (Eox + 4.80)[eV], E(LUMO) = - (Ered + 4.80)[eV], where Eox and Ered are the oxidation and reduction onsets.

Figure 4. DSC thermograms showing three heating-cooling cycles for the octyl (1) and hexyl (2) phthalimide derivatives

Figure 5. Current–voltage curves of as-casted and thermally annealed devices of: A. Octyl-phthalimide derivative (1)/PC61BM blend. B. Hexyl-phthalimide derivative (2)/PC61BM blend

Figure 6. AFM topographic images of as-cast blend films of donor/fullerene blends. A. 20 µm × 20 µm. B. 5 µm × 5 µm

Table 1. Solution cyclic voltammetry estimation of HOMO and LUMO of the octyl (1) and hexyl (2) phthalimide derivatives and DTS(FBT-Th2-Hexyl)2

Table 2. Average photovoltaic parameters of octyl-phthalimide derivative (1)/PC61BM, hexyl-phthalimide derivative (2)/PC61BM BHJ OPV devices

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Highlights:

- Synthesis of phthalimide end-capped derivatives of DTS(FBT-Th-Pth-Hexyl)2

- Compounds exhibit higher solubility and deeper HOMO levels

- Solar cell devices can achieve higher open circuit voltages

- Performance limited by unfavorable morphology with large domain phase separation